A Case Study of the Budget Planning Bias

Transcription

1 Statement of Douglas Holtz-Eakin Director The Pension Benefit Guaranty Corporation: Financial Condition, Potential Risks, and Policy Options before the Committee on the Budget United States Senate June 15, 2005 This statement is embargoed until it is delivered at 10 a.m. (EDT) on Wednesday, June 15, The contents may not be published, transmitted, or otherwise communicated by any print, broadcast, or electronic media before that time.

2

3 Chairman Gregg, Senator Conrad, and Members of the Committee, I welcome this opportunity to discuss with you the important issues of pension insurance and pension plan funding. Those issues are important to workers and firms that participate in defined-benefit pension plans as well as to U.S. taxpayers. I will focus my remarks on four themes: # PBGC s costs can be usefully divided into prospective and so-called sunk costs. Changes in policy can reduce or avoid losses that have not yet occurred, whereas the losses from plans that are already terminated or are in the process of terminating can only be paid. # Recent experience shows that lack of clarity in financial information about an insurance program can effectively obscure rising costs and delay policy responses. # Premiums and funding rules that reflected the risks and costs that various plans imposed on PBGC could match costs with behavior and provide incentives for firms to reduce those costs. # Policies that reduced costs to PBGC could also reduce federal revenues. Those two effects should be taken into account in assessing policy changes to avoid simply transferring PBGC s costs to the revenue side of the budget. Accrual Accounting and Exposure to Underfunding At present, the underfunding of defined-benefit pension plans is a pervasive and sizable phenomenon. PBGC estimates that the vast majority of plans are currently underfunded to some degree. The agency s best estimate of total underfunding (on a termination basis) among all insured plans is $600 billion $450 billion for single-employer plans and $150 billion for multiemployer plans. 1 Fortunately, most underfunded plans are not likely to be terminated because they are sponsored by financially healthy firms. Therefore, PBGC assesses the amount of underfunding among plans for which the agency considers default reasonably 1. By law, the funding rules and insurance system treat pension plans sponsored by a single employer differently from those sponsored by more than one firm, which are referred to as multiemployer plans. Although both types of plans are experiencing similar problems, PBGC underwrites much more liability for single-employer plans, and as a result, most efforts at pension reform concentrate on them.

4 possible. In fiscal year 2004, PBGC estimated its exposure to claims from such plans at $96 billion. 2 Even without taking those prospective costs into account, PBGC is already in a deep fiscal hole. At the end of 2004, the agency was reporting a negative net financial position of $23.5 billion. PBGC s net financial position essentially measures how the resources available to the agency at a given point in time compare with the pension obligations from plans that have already been terminated as well as claims from plans whose termination in the near future PBGC considers probable. Of course, all estimates of underfunding are just that: estimates. As such, they are sensitive to projections about interest rates, future returns on assets, retirement ages, and life expectancies. A shift in those factors could have a substantial effect on projections of underfunding. Economic Costs Under the Employee Retirement Income Security Act of 1974 (ERISA), PBGC is not backed by the full faith and credit of the U.S. government and has no authority to change the premium structure or call on general revenues to pay benefits. The resources at its disposal are premiums, the assets from terminated plans, and investment income from its accrued assets. Therefore, if PBGC exhausted all of its holdings, as is projected to occur under current law, it would have to rely almost entirely on its ongoing stream of premiums to cover its expenses. That circumstance would in turn necessitate a drastic reduction in benefits perhaps in excess of 90 percent. As a practical matter, however, the public probably views the pension insurance system as carrying an implicit federal guarantee. Consequently, many observers expect that if PBGC became insolvent, the Congress would feel compelled to provide direct assistance from general revenues. How extensive is that implicit guarantee? Over the past 18 months, the Congressional Budget Office (CBO) has been analyzing the economic costs that PBGC s insurance represents for taxpayers if the implicit guarantee is honored. That work is still in progress; currently, however, CBO estimates that the economic costs to the public of PBGC s insurance for single-employer plans 2. That reasonably possible termination category includes primarily plans sponsored by firms that the financial markets consider to be experiencing some financial distress indicated by credit ratings below investment-grade but that are not already included among plans whose termination PBGC rates as probable. 2

5 (including sunk as well as prospective costs and subtracting premium collections) total $71 billion for the upcoming decade and $91 billion over the next 20 years. Those figures describe the estimated net present value of the financial resources that PBGC will be transferring to sponsors of and participants in defined-benefit pensions. They are also the estimated prices that the government will have to pay to private insurers bidding in competitive markets to take on the obligations that PBGC will assume during those periods under the current premium structure and funding rules. Thus, they reflect both the costs that are likely to be incurred in coming years and the risk that those costs could be even greater than anticipated. In considering how to address the economic costs that PBGC s insurance represents, it is critical to distinguish between costs that have already been incurred and costs that are likely to be incurred in the future. At the end of 2004, PBGC had accumulated losses in the single-employer program of $23.3 billion for plans that had been terminated or plans whose termination the agency regarded as probable. Those sunk costs cannot be avoided, and policy decisions can determine only who will bear them. But changes in policy can reduce prospective economic costs, which, according to CBO s current estimates, are $48 billion for the next 10 years and $68 billion for the next 20 years. PBGC s Cash Resources Increased pressure to provide additional resources to PBGC may arise once the shortfall between liabilities and assets from terminated plans begins to register as a deficit in the agency s annual bottom-line measurement of its cash flow. Under PBGC s current premium structure and funding rules and the assumptions of CBO s current economic forecast, the agency will soon start running cash deficits, which will continue for the foreseeable future. In CBO s projections, the combination of growing obligations for benefits and level income from premiums causes the agency s on-budget fund to be exhausted in about No precedent exists for how PBGC would proceed if its on-budget fund became insolvent. However, CBO s expectation is that the agency would cover its costs by increasing the percentage of benefits and other expenses being paid out of its nonbudgetary trust fund. (By doing so, PBGC would essentially be providing less insurance protection to future recipients in terminated plans than it provides to current recipients.) CBO does not formally estimate the value of the assets held by 3. The resources available to pay for PBGC s costs are divided between two funds: an on-budget fund for receipts of premiums and outlays for benefits and administrative costs, whose transactions since 1980 have been included in federal budget totals; and a nonbudgetary trust fund, in which the assets of terminated plans are held until used to help pay benefits. 3

6 the trust fund. There is a significant likelihood, however, that all of PBGC s assets will be exhausted within the next 20 years. Alternative Budgetary Treatments The potential liability that pension insurance represents for taxpayers might be made more visible by changing PBGC s budgetary treatment. Today, that treatment focuses on the cash inflows (primarily from premiums, interest income, and transfers from the nonbudgetary trust fund) to the agency s on-budget account, which are then subtracted from the federal government s outlays for the pension benefits that PBGC pays and for its administrative expenses. That treatment delays, often for decades, the budget s recognition of PBGC s insurance claims from when they are realized at a plan s termination to when benefits are paid. As a consequence and despite large losses PBGC s budgetary position has helped reduce the federal deficit in every year except That kind of budgetary treatment is neither designed to indicate nor suited to describe the expected risk and magnitude of losses in the pension insurance system. A budgetary treatment that better indicated the full costs of pension insurance would have the following attributes: # Timeliness of Recognition. The budget should reflect costs when the government incurs the obligation to pay them. 4 Although sunk costs must be recorded and paid, it is the costs that are being incurred during a budget period that are the focus of policymakers decisions. One possibility would be to include the losses that PBGC incurs on pension plans when those plans are terminated rather than when the benefit payments are actually made. Of course, under current law, the extent of the government s commitment to pay benefits is restricted to the resources available to PBGC from premiums, the assets of terminated plans, and whatever it can recover from plans sponsors. # Market-Value Basis. The best way to assess the cost of an insurance program is by using market prices to value the risk associated with it. For PBGC, the market price of risk is significant because the events that are most likely to precipitate a transfer of pension liabilities to the agency (including low investment returns, high rates of financial distress, and low interest rates) generally occur when the market value of all assets has dropped. 4. President s Commission on Budget Concepts, Report of the President s Commission on Budget Concepts (October 1967), p

7 The current budgetary treatment of PBGC recognizes the inflow of premium collections and the outflow of benefit payments during the budget period, but it does not take into account the value of the claims arising under the insurance and thus does not manifest the attributes outlined above. CBO is currently exploring budgetary alternatives that might better reflect those qualities. For example, if policymakers wanted the federal budget to reflect PBGC s prospective economic costs rather than its current cash flows, one possibility would be to treat those net prospective costs as the agency s baseline costs. Future year budgets could then recognize alterations in the value of the insurance as a result of changes in law, regulations, or such variables as insured liabilities and interest rates. Similar to the way that loan programs are treated under credit reform accounting, those changes in costs might be considered either reestimates (the result of unexpected economic changes) or modifications (the result of policy changes). That treatment would capture the magnitude of future claims from unfunded insured pensions, but it would also depart from standard budgetary treatments by including costs for which the government is not currently liable. In addition, unlike credit reform accounting, the values presented could include the cost of market risk. Another possibility would be to recognize as budgetary costs the unpaid fairmarket value of premiums for PBGC s insurance that is, estimates of the annual premiums required to reduce to zero the net economic costs of the insurance that PBGC provides. Such unpaid premiums could be compared with the premiums that are expected to be paid by plans sponsors, and the difference could be shown as the budgetary costs of PBGC. (As with the option above, this treatment would also depart from standard budget presentations by displaying costs for which the government was not currently liable and by including the cost of market risk.) An alternative approach would be to transfer PBGC to private owners. That step would probably accelerate the recognition of sunk costs in the budget because PBGC s current deficit would have to be covered, presumably through appropriated funds, before a private entity would be willing to assume the agency s obligations. In addition, a private owner might require either an annual or lump-sum payment from the federal government to continue to operate the insurance program under its current premiums and funding rules. Because PBGC s insurance is mandatory for defined-benefit pension plans, the government would probably remain involved in regulating the terms of the insurance which raises the question of how much risk and responsibility the government could effectively transfer to private owners. Nevertheless, the risk to taxpayers would most probably be less under such an arrangement than it is under current policy. 5

8 Policy Proposals Under Consideration Defined-benefit pensions are a form of employee compensation. The objective of policies directed toward improving pension insurance is to provide a framework to support payment of that compensation despite the potential for adverse economic events affecting the sponsoring firm, particular industries, or the economy as a whole between the time when the compensation is earned and the time when the pension benefits are paid. In addition, it would be desirable for such policies to support or at least not impede any economic restructuring that changes in competitive pressures might induce. Two broad policy avenues are available: sponsoring firms could be required to accumulate more resources to pay promised benefits, or they could pay the full cost of purchasing insurance from PBGC to provide the necessary resources. Under both types of policies, it would be important for firms to face the full cost of their decisions about compensation through rules that enforced adequate funding and insurance that was appropriately priced to reflect the risk of losses particular firms posed and for there to be sufficient transparency for markets to enforce those incentives. Alter the Premium Structure The underpricing of PBGC s insurance that is, the current premium structure is a key factor in the agency s present financial difficulties. Premium revenue is the only source of income available to PBGC to cover the shortfall between the liabilities of terminated plans and the value of their assets. CBO expects that under current law, premium income will remain relatively flat at around $1 billion annually whereas benefit payments resulting from both past and future claims will rise from about $3.5 billion this year to more than $10 billion in A contributing factor to that pattern is that the premium rate paid by sponsors of multiemployer plans has remained constant since 1988, and rates for the two types of premiums charged for single-employer plans have not changed in more than a decade. (One of those premiums is an amount levied per plan participant; the other is calculated on the basis of a plan s underfunding.) 5 The rates for the premiums are set by statute, and PBGC cannot adjust them, as most insurance providers can, for the losses that past events lead it to expect. Raising premiums would require sponsors to pay a larger share of PBGC s economic costs. To cut federal costs to zero through higher premiums alone would require a fivefold increase in the agency s receipts from premiums. Those higher premiums might be manageable for well-funded plans, which currently pay 5. The premium levied on underfunding does not always work as intended. Because of loopholes in the premium rules, many plans that are underfunded are not actually required to pay premiums on their underfunding. 6

9 only a flat charge of $19 per year per participant for insurance. Firms whose plans are significantly underfunded, however, pay not only the flat rate per participant but also a charge of $9 per $1,000 of underfunding. (A hypothetical firm with 1,000 participants and $50 million in underfunding pays premiums of $469,000 per year, of which $450,000 is the charge for underfunding.) Therefore, for some firms, an increase in premiums could be significant perhaps to the point of causing them to adjust the form and amount of compensation that they offer. An alternative to a proportionate increase in premiums for all plans sponsors would be to make premiums more sensitive to the risk that various plans pose for PBGC. Although the extra charge for underfunding currently provides some adjustment based on risk, varying premiums on the basis of risk could reduce the current cross-subsidies from low-risk sponsors and plans to high-risk ones. Some risk-adjusted premiums could also strengthen incentives for firms to reduce risk which could lower the premium rate required to achieve any given level of net costs. Under a risk-based approach, premiums would be higher for sponsors that were more likely to encounter financial distress and whose plans tended to be more deeply underfunded at termination. For example, premiums could vary with the volatility of the market value of a firm and its pension assets, the ratio of the firm s liabilities to its equity (leverage), or the firm s credit rating. The resulting range of premiums would be substantially wider than it is under current policy because risk varies significantly among plans. Another important correlate of plans risk that could provide a basis for adjusting premiums is the ratio of a pension plan s assets in stocks to its total assets. Plans sponsors appear to prefer to hold a large proportion of their assets as equities because, historically, stocks have yielded higher average returns (but at greater risk) than have bonds. If those higher returns are realized, the risk premium that they represent serves to reduce the cash contributions that a sponsor must make to its plan to fund the pension benefits it has promised. Of course, investments in equities entail the risk that the stock market will do poorly and the plan will become underfunded. Indeed, plans that hold a large proportion of common stocks, rather than high-quality bonds or other fixed-income securities, exhibit more volatility in the value of their assets than do plans that hold more debt securities. Plans with a large share of stocks are thus at greater risk of underfunding when their sponsors encounter financial distress. Because PBGC s costs vary more closely with plans liabilities than they do with the number of participants in plans, the current premium structure does not reflect the chances that PBGC will take on particular claims. The current per-participant charge tends to result in lower premiums per dollar of insured liabilities for firms with a higher proportion of older or high-wage employees compared with firms 7

10 whose workforce is predominantly younger or lower paid and therefore has few accumulated pension benefits. At the current rate of $19 per participant, those effects may be small, but if rates were raised to be fair, on average, the effects on firms behavior could be significant. Another issue relevant to the pricing of pension insurance is how premiums should be changed to reflect past versus future claims against PBGC. The estimated shortfall for past claims as well as some imminent losses is $23.5 billion. CBO has estimated that the value of PBGC s insurance over the next two decades will be $68 billion. That is, the agency may soon be taking on billions of dollars more in claims. If premiums were set so as to lessen or eliminate the agency s accumulated deficit as well as to accurately reflect its exposure to future claims, ongoing sponsors of plans would be charged more than actuarially fair rates. That kind of a system might lead some sponsors of well-funded plans to freeze or terminate their plans which would actually worsen PBGC s finances by reducing its premium collections. In considering how to finance pension insurance in coming years, it would be useful to address the following as separate issues: (1) how to price pension insurance to cover future risks and provide the proper economic incentives to firms in managing their pension plans and (2) how to pay for losses that have already been incurred. Change Funding Rules and Reporting The current rules governing pension funding were intended to ensure that firms contributed adequate resources to pay promised benefits by the time the benefits came due and to provide firms with some flexibility as to when and how they made those contributions. However, certain features of those rules may have led to systematic underfunding among a number of defined-benefit plans. Many firms whose pension plans were recently taken over by PBGC used those features to make small or no contributions in the years leading up to the plans termination at which point they presented PBGC with billions of dollars in claims. A number of options are available for strengthening pension funding rules, and all involve trade-offs that might make them more or less attractive to a particular stakeholder in the pension system. Instead of attempting to enumerate them all, the discussion that follows broadly describes several approaches and spells out some general principles that might guide reform. Price to the Market. Under the current set of funding rules, plans liabilities are assessed not according to market values but on the basis of a four-year weighted average of interest rates, a practice known as smoothing. The current actuarial valuation of assets relies on a smoothing technique as well. (Another example in which assets and liabilities are not priced to market involves credit balances from previous-year contributions that exceeded the minimum funding requirement, 8

11 which are calculated without regard to changes in market values.) When markets (and rates) are changing rapidly, the funding ratios (assets to liabilities) that plans report under the current funding rules may be markedly different from the ones that would result from calculations that used current market values. In recent years, such discrepancies have led plans to appear better funded than they actually are. (Of course, in a different economic environment, the reverse could be true.) Valuations of pension assets and liabilities that were more closely linked to current economic conditions would provide a more accurate picture of plans funding status and provide a better base on which to set funding requirements. Some observers have also suggested that using current market values for liabilities and assets would encourage plans to invest their assets in a way that better matched the duration of their liabilities with the projected income from those assets. That approach would help insulate plans from financial fluctuations and thus moderate the volatility of required contributions. Match Characteristics of Assets to the Nature of Liabilities. Plans are required to pay for most pension benefits as those benefits are accrued, but they have great leeway in deciding how to invest their accumulated pension assets. As noted earlier, most plans attempt to take advantage of the opportunity to realize an equity premium by investing in stocks rather than bonds. (On average, about 70 percent of the pension assets of publicly traded companies are invested in equities, and most of the rest is invested in bonds or held as cash.) Although, historically, stocks have yielded higher average returns than bonds over the long run, they are also more volatile, which makes them unsuited to financing pension benefits that will come due in the short term. The combination of low stock values and low interest rates over the past several years helped create a large amount of underfunding, which sponsors are now being required to make up. Investment behavior by sponsors that more closely matched the characteristics of investments with the expected duration of liabilities would have enabled plans to avoid much of the underfunding they now face. Creating incentives in the funding rules to encourage plans to more closely match the type of assets they hold to the duration of their liabilities would lead to fewer large swings in funding levels and put PBGC at less of risk of having to absorb sudden increases in pension shortfalls. Consider All Relevant Costs. The current funding rules do not take full account of all costs that a pension plan may represent. For instance, some plans provide lump-sum payments to participants if a particular facility shuts down. In addition, plans that are nearing termination often experience a sudden increase in costs as many employees take early retirement. Although not every plan will experience the surge in costs associated with shutdown benefits or a sudden flurry of early retirements, those events can substantially increase PBGC s costs if a plan is terminated. Therefore, considering ways to measure liabilities that included all relevant contingent liabilities along with the likelihood of incurring those costs would be prudent. 9

12 Make Risk Part of the Equation. Just as with premiums, it would be possible to link certain funding requirements with the risk that a plan will terminate. Data about plan terminations suggest that so-called distress terminations are strongly linked to the credit ratings of plans sponsors. According to the Government Accountability Office, of the largest 41 claims in PBGC s history for which a credit rating was known, 39 plans were sponsored by firms whose credit was rated as speculative at least three years prior to their plans termination. Those data suggest that one way to help prevent large claims for PBGC would be to structure the funding rules to minimize underfunding in plans sponsored by less creditworthy firms. The difficulty with that approach, however, is that firms with lower credit ratings often exhibit weaker cash flows than firms with higher ratings and have limited access to capital in the credit and equity markets. Improve Transparency. Markets work best when full information is available to participants. The current pension system does not do a very good job of providing the kind of information that is helpful to investors and plans participants as well as to policymakers and taxpayers. Funding levels are measured in different ways for different purposes, and information about potential underfunding that is filed with PBGC and other government agencies (such as the Internal Revenue Service) often lags years behind. The lack of transparency can cause investment markets to undervalue sponsors costs for providing pension benefits; it may also lead workers to underestimate the likelihood that their promised pensions will not be delivered in full. Regulatory changes that led to greater transparency would help stakeholders (including plans participants and investors) to evaluate whether sponsors were meeting their obligations. Although the effect would be difficult to quantify, that increased level of scrutiny could discourage sponsors from underfunding their plans or committing their firms to obligations that could not be kept. Improve Flexibility. At various times in ERISA s history, the law has limited the ability of sponsors to effectively overfund their plans. Those restrictions have been reduced over time, mostly in an effort to limit the losses of federal revenue that those contributions may represent. (Contributions to the plans are tax-deductible to the sponsoring companies to the extent that they are profitable enough to owe taxes.) The evidence is mixed as to whether plans will actually contribute more than is required, even during good economic times. However, the argument is often made that allowing sponsors to effectively overfund their plans can provide them with a buffer in the event of an economic downturn. The Administration s Proposal The Bush Administration has proposed several changes in the defined-benefit pension system to reduce its financial shortfall and increase transparency. 6 In 6. Details are available at 10

13 general, the Administration would raise premiums and permit further adjustment of them for risk, change the measure of plans liabilities and funding requirements, and increase public disclosure of plans funding status. The sponsors of plans would also be permitted to fund the plans liabilities at higher levels during good economic times (without the loss of tax benefits) as a buffer against underfunding during less prosperous periods and to use a higher discount rate to calculate plans liabilities. 7 Most of those changes are consistent with the objective of reducing the federal costs of pension insurance. Reconciliation and PBGC s Deficit The reconciliation instructions associated with the Concurrent Budget Resolution for Fiscal Year 2006 directed that the Senate Committee on Health, Education, Labor, and Pensions reduce outlays for mandatory programs within its jurisdiction by $13.6 billion over the period. The resolution does not specify the amounts that must come from each program of mandatory spending within the committee s jurisdiction, but the bulk of that spending covers higher education programs and PBGC. It has been widely reported that roughly $6 billion to $7 billion of the savings is expected to come from PBGC. To put that target in perspective, under current law, PBGC s gross outlays will total $34 billion over the five-year period, in CBO s estimation, and premiums will bring in $7 billion. One option for meeting the reconciliation bill s target would be to roughly double total premium income to around $14 billion over the period. Such an increase would be enough to erase PBGC s on-budget cash deficit over that period and avoid exhaustion of the assets in its on-budget fund during the budget window. However, the increase in premiums would not be sufficient to cover the $48 billion in economic costs that CBO estimates the agency is likely to incur over the next 10 years. Reconciliation targets are a measure of the agency s annual cash flows, whereas PBGC s economic costs are the net present value of its insurance claims (that is, net of premiums) over the next decade. Achieving the reconciliation target from an increase in flat-rate premiums alone would require that the current per-participant charge be more than tripled, from $19 to nearly $60. Such a policy could generate the needed $7 billion in additional cash receipts over five years and about $14 billion over 10 years, but the premium increase would reduce the economic cost of providing insurance by less than $7 billion over the same 10 years, to $41 billion. 7. The present value is a single number that expresses a flow of current and future income (or payments) in terms of an equivalent lump sum received (or paid) today. Market interest rates are the basis of the discount rate used to calculate the net present value of plans liabilities. 11

14 Another approach would be to increase the variable-rate premium, which is charged on the amount of underfunding in each plan. Tripling the variable-rate premium would generate about $6 billion in additional cash receipts over the fiveyear period. It would also reduce PBGC s economic costs by nearly $18 billion, but it could cause some sponsors to terminate or freeze their plans. 8 (Meeting a fiveyear target through the variable-rate premium is further complicated by a lag in the collections.) Changes to the funding rules that reduced PBGC s future benefit payments by reducing the size of the claims it was likely to take on could significantly reduce the agency s long-term economic costs but would do little to help meet the reconciliation targets. That circumstance results because any reduction in underfunding among active plans would occur over a number of years and its effect on claims in the short run would be relatively small. The effect of reduced underfunding on PBGC s outlays over the five-year period would be even smaller. Implications for Revenues Changes to the pension funding rules that affected the mandatory spending covered by the reconciliation instructions to the Senate Committee on Health, Education, Labor, and Pensions would also have an impact on federal revenues. Changes that required sponsors to increase contributions to their plans, which are tax-deductible, would result in firms redirecting some resources slated for other purposes. Presumably, many of those resources would be directed away from taxable forms of spending. Thus, efforts to limit PBGC s future claims by reducing underfunding within the pension system would also tend to reduce federal revenues. However, because defined-benefit pensions are a form of tax-deferred compensation, some tax revenue would eventually be realized from the additional contributions (although far outside the 10-year budget window) when those contributions were paid out as benefits that otherwise would not have been paid in full. 8. Neither the estimated receipts nor the estimated economic costs currently reflect any response on the part of affected sponsors. 12